Bugzilla – Full Text Bug Listing
| Summary: | Agent Controller fails to start up due to shared memory TL error | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | z_Archived | Reporter: | Samson Wai <samwai> | ||||||||||
| Component: | TPTP | Assignee: | Igor Alelekov <igor.alelekov> | ||||||||||
| Status: | CLOSED FIXED | QA Contact: | |||||||||||
| Severity: | critical | ||||||||||||
| Priority: | P1 | CC: | igor.alelekov, jkubasta, karla.callaghan, paulslau | ||||||||||
| Version: | unspecified | Keywords: | plan | ||||||||||
| Target Milestone: | --- | ||||||||||||
| Hardware: | PC | ||||||||||||
| OS: | Linux | ||||||||||||
| Whiteboard: | closed460 | ||||||||||||
| Attachments: |
|
||||||||||||
|
Description
Samson Wai
This should be set to 4.4i1 P1. Please set priority to P1. Please set priority to P1 Samson, could you repeat AC launching as superuser (root)? Does the problem remain? It starts only if I log on as root. It should not require root to start AC just like the ChkPass utility I used for verifying user passwords. After AC get terminated it leaves the shared memory block (acbuffer) in the system (Linux), and current user is marked as owner of this memory. Another user can't launch AC due to permission restriction in creating acbuffer. The problem could be fixed by clearing (destroying) of the shared memory on AC termination. Set to P1. Created attachment 58661 [details]
RASharedMemory patch
Created attachment 58662 [details]
transport components (transportSupport and SharedMemTL) patch
The patch releases acbuffer shared memory block on AC termination. The patch is devided on two parts: - RASharedMemory - affects RAC code - transport - affects AC code Samson, please review RASharedMemory patch Joanna, who will review second part (transport) of the patch? Created attachment 58753 [details]
modifed patch for transport components
Modified SharedMemTL.dsp (windows project file) has been added to the patch.
Igor, would you please ask Randy or Kevin to review the patch for the transport components? Hi Igor, the RAC-side fix looks good to me. It seems that what you are trying to do is to expose the stop flusher routine from the RAC code base and use it in the AC code base. The RAC itself is not using this newly exposed function as all. Am I correct? On the other hand can this fix handle the case where there was a crashed AC leaving the shared memory uncleared? It seems not... Will we be hit by the same problem then? > Hi Igor, the RAC-side fix looks good to me. It seems that what you are trying > to do is to expose the stop flusher routine from the RAC code base and use it > in the AC code base. The RAC itself is not using this newly exposed function as > all. Am I correct? Yes. > On the other hand can this fix handle the case where there was a crashed AC > leaving the shared memory uncleared? It seems not... Will we be hit by the same > problem then? Currently any normal AC shutdown leaves uncleared shared memory block acbuffer. This doesn't allow launch AC by other users. The patch fixes this issue. As for crashed AC - additional investigations are required. Hi Igor. I would recommend addressing the crash scenario as well when fixing this bug. This will prevent users from hitting the same problem if they choose to run "kill -9" on the Agent Controller. (In reply to comment #15) > Hi Igor. I would recommend addressing the crash scenario as well when fixing > this bug. This will prevent users from hitting the same problem if they choose > to run "kill -9" on the Agent Controller. Yes, it is important. But it seems that the patch, fixing the normal termination scenario could be aplied. And additional investigations of the crash scenario could be done bit later. *** Bug 175264 has been marked as a duplicate of this bug. *** The attached two patches have been committed to CVS. Igor please mark this one as closed. Please also open a new bug to investigate to see if we can handle the crash scenario for 4.4. If we cannot fix that then we will need to add a readme entry telling user what to clean up after an AC crash. Thanks. resolving as fixed, new bug #177153 is opened to investigate the AC crash scenario Created attachment 60903 [details]
patch for the SharedMemTL project file
Hi Samson, Review, please, the patch for the SharedMemTL project file. It appends necessary link to the Release configuration. The same link in the Debug configuration already done by the previous patch. Igor. It is now checked in. Thank you. Resolving as fixed. As of TPTP 4.6.0, TPTP is in maintenance mode and focusing on improving quality by resolving relevant enhancements/defects and increasing test coverage through test creation, automation, Build Verification Tests (BVTs), and expanded run-time execution. As part of the TPTP Bugzilla housecleaning process (see http://wiki.eclipse.org/Bugzilla_Housecleaning_Processes), this enhancement/defect is verified/closed by the Project Lead since this originator of this enhancement/defect has an inactive Bugzilla account and considered to be fixed. If this enhancement/defect is still unresolved and reproducible in the latest TPTP release (http://www.eclipse.org/tptp/home/downloads/), please re-open. |